Bipolar I disorder (BPI) is a severe neuropsychiatric disorder that afflicts approximately 1% of most populations. Pivotal neurobiological underpinnings of BPI have yet to be unraveled but family, twin and adoption studies carried out over the past half century indicate that hereditary factors underlie at least 80% of disease variance. Recent genome wide association (GWA) studies of BPI indicate that common variants with small effect sizes explain no more than 3% of the genetic liability. The missing heritability is thought to be explained by many common variants with very small effect sizes and/or rare variants with relatively large effect sizes. Families with a large percentage of affected individuas in more than one generation are likely to contain coding or regulatory variants with large effect sizes. This study focuses on identifying rare susceptibility variants in two multigenerational BPI pedigrees of European ancestry, each containing at least 7 BPI cases across at least two generations, ascertained through the NIMH Genetics Initiative. These families have been recently genotyped, and there are linkage data that will help prioritize the search for rare variants. We have selected 3 cases for whole genome sequencing in each family. The work will be contracted out to a private biotechnology company at a cost of no more 4.0K per genome. We will prioritize all rare variants shared by the 3 BPI cases in the exome and proximal regulatory and promoter regions. A novel analytic program, ANNOVAR, will be used for prioritizing rare variants. We will check for co-segregation of potential disease predisposing rare variants using all affected family members with DNA available, and for the variants that remain, we will apply for funding to carry out gene-based case control studies in unrelated affecteds and controls.